Link Dump - jejjohnson/research_journal_v2 GitHub Wiki

Idea - Process-Oriented ML for Constraining Climate Models

• StoryLines Methodology from ML Perspective - AI4Good | Paper I | Paper II | Climate Invariant Machine Learning
• Missing I - Latent Variable Model + Uncertainty Quantification + Sensitivity Analysis
• Missing II - Dynamical Model…

---

SWOT DATA.

• Link to webpage with example access points - EarthAccess
• also includes other links to other datasets - ECCO, GHRSST, OPERA, Sentinel-6, SMAP, S-MODE

GLORYS

• CMEMs https://github.com/meom-group/download-datasets/tree/main/cmems

---

xarray & ml datasets

• xarray-tensorstore
• neuralgcm code
• xarray + zarr + dataset + cloud

---

Ensemble Kalman Filter

• Iterative Ensemble Smoother

---

Emulators

Spatial Operator: Transformer

• Stormer
• xiHE - Ocean Model

Spatial Operator: Neural Operator

• OceanNet

Spatial Operator: Graphs

• Neural-Lam

Spatial Operator: Hybrid

• Neural-GCM

Time Stepper

• RollOuts for Neural Operators

Terminal Warrior

webman

• cat —> bat
• pip —> uv
• grep—> ripgrep
• htop —> zenith
• fswatch —> watchexec
• ls —> eza
• cd -> zoxide
• joshuto

---

Spatial Operators

Graphs

• xarray to networkx - Example

---

Research

Ocean

• A neural network-based method for merging ocean color and Argo data to extend surface bio-optical properties to depth: Retrieval of the particulate backscattering coefficient
• An Alternative to Static Climatologies: Robust Estimation of Open Ocean CO2 Variables and Nutrient Concentrations From T, S, and O2 Data Using Bayesian Neural Networks
• Recent variability of the global ocean carbon sink
• A neural network-based estimate of the seasonal to inter-annual variability of the Atlantic Ocean carbon sink